This post is 500 words and takes about 2 minutes to read.

Most of us presume that cities have more police when there is more crime.  We also assume that more people in prison is a result of more crime.  But two recent studies say that may not be the case in Canada at all.

Studies by McGill professor Jason Carmichael and colleagues have concluded that the best predictor of having more police per capita, and of having more people imprisoned per capita in Canada is not crime rate, but higher numbers of Aboriginal people and visible minorities.

Let’s restate that for clarity: provinces with more people in federal prison, and Canadian cities with more police per capita tend to be those with higher populations of visible minorities and Aboriginal people, not those with more crime.

However the researchers also found that the political stripe of the provincial government did not predict either police numbers or incarceration rates.

The studies

These conclusions come from studies published in 2015 by researchers from McGill, Harvard, and Cleveland State universities.  The researchers used a sophisticated statistical technique called multiple regression to compare the relative importance of a whole range of factors on police numbers and on numbers of people in federal prison.  The study on police used data on Canadian cities with more than 100,000 people from 1996 to 2006.  The study on imprisonment used provincial data from 2001 to 2010.  Among the factors the researchers considered were income inequality, the percentage of poor and unemployed, the number of young people, and the size of the city.  None of these showed a relationship to policing or jailing.  Nor was there a strong relationship between the number of people charged with a crime and the number in prison.  The size of the police budget as a proportion of the total city budget was also not related to police per capita.

The researchers say that these data are consistent with the idea of policies driven by the perception of ‘ethnic threat’.  Many U S studies have found that numbers of visible minorities have a much stronger correlation with police presence and numbers jailed than do other factors such as actual crime rates or poverty rates. That is, more use of punitive power is associated with fear, and fear is associated with people who look different. As they put it, “majority group members will use their disproportional influence over the criminal law and the criminal justice system to support more punitive policies and practices to control those groups perceived as threatening”.

However previously nobody had tested this idea using Canadian data, and there are, of course, many important differences between Canada and the United States in ethnic composition of the population, operation of the criminal justice system, and in the politics around crime.

These findings are consistent with other Canadian data (see earlier post) showing differences in criminal justice elements related to ethnicity and skin colour.



The two articles on which this post is based are available here (prisons) and here (police).  There is also a McGill University press release on the studies.

Jason Carmichael is at McGill University. Stephanie Kent is at Cleveland State University.  Roland Neil is at Harvard University.